view scripts/image/hsv2rgb.m @ 20404:131ce8cfaa80

Relax input in functions that convert between colorspaces (bug #45456) * scripts/image/hsv2rgb.m, scripts/image/ntsc2rgb.m, scripts/image/rgb2hsv.m, scripts/image/rgb2ntsc.m: remove all input check and leave it up to new private functions handled by all. This adds support for ND images, drops check for values in the [0 1] range, allows integer images, and allows sparse matrices. Also adjust tests and add extra ones. * scripts/image/private/colorspace_conversion_input_check.m, scripts/image/private/colorspace_conversion_revert.m: all of this functions handle input check in the same way. In the same way, they need to prepare the output in the same way based on what preparation was done during input check (transforming image into colormap). So we create two new private functions to avoid repeated code. All code was adapted from hsv2rgb.
author Carnë Draug <carandraug@octave.org>
date Sun, 19 Jul 2015 17:41:21 +0100
parents 84ca63c8a038
children
line wrap: on
line source

## Copyright (C) 1999-2015 Kai Habel
## Copyright (C) 2015 Carnë Draug
##
## This file is part of Octave.
##
## Octave is free software; you can redistribute it and/or modify it
## under the terms of the GNU General Public License as published by
## the Free Software Foundation; either version 3 of the License, or (at
## your option) any later version.
##
## Octave is distributed in the hope that it will be useful, but
## WITHOUT ANY WARRANTY; without even the implied warranty of
## MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
## General Public License for more details.
##
## You should have received a copy of the GNU General Public License
## along with Octave; see the file COPYING.  If not, see
## <http://www.gnu.org/licenses/>.

## -*- texinfo -*-
## @deftypefn  {Function File} {@var{rgb_map} =} hsv2rgb (@var{hsv_map})
## @deftypefnx {Function File} {@var{rgb_img} =} hsv2rgb (@var{hsv_img})
## Transform a colormap or image from HSV to RGB color space.
##
## A color in HSV space is represented by hue, saturation and value
## (brightness) levels in a cylindrical coordinate system.  Hue is the
## azimuth and describes the dominant color.  Saturation is the radial
## distance and gives the amount of hue mixed into the color.  Value is
## the height and is the amount of light in the color.
##
## The input can be both a colormap or RGB image.  In the case of floating
## point input, values are expected to be on the [0 1] range.  In the case
## of hue (azimuth), since the value corresponds to an angle,
## @code{mod (h, 1)} is used.
##
## @example
## >> hsv2rgb ([0.5 1 1])
## @result{} ans = 0 1 1
##
## >> hsv2rgb ([2.5 1 1])
## @result{} ans = 0 1 1
##
## >> hsv2rgb ([3.5 1 1])
## @result{} ans = 0 1 1
## @end example
##
## Output class and size will be the same as input.
##
## @seealso{rgb2hsv, ind2rgb, ntsc2rgb}
## @end deftypefn

## Author: Kai Habel <kai.habel@gmx.de>
## Adapted-by: jwe

function rgb = hsv2rgb (hsv)

  ## Each color value x = (r,g,b) is calculated with
  ## x = (1-sat)*val+sat*val*f_x(hue)
  ## where f_x(hue) is a piecewise defined function for
  ## each color with f_r(hue-2/3) = f_g(hue) = f_b(hue-1/3).

  if (nargin != 1)
    print_usage ();
  endif

  [hsv, cls, sz, is_im, is_nd, is_int] ...
    = colorspace_conversion_input_check ("hsv2rgb", "HSV", hsv);

  h = hsv(:,1);
  s = hsv(:,2);
  v = hsv(:,3);

  ## Values of Saturation and Value should also be in the [0 1] range.  With
  ## the exception of hue, values outside this range don't make any sense
  ## in a cylindrical coordinate system but we must return something for
  ## Matlab compatibility.  User case is when a function returns an hsv
  ## image just slightly outside the range due to floating point rounding
  ## errors.

  ## Prefill rgb map with v*(1-s)
  rgb = repmat (v .* (1 - s), 1, 3);

  ## red = hue-2/3 : green = hue : blue = hue-1/3
  ## Apply modulo 1 to keep within range [0, 1]
  hue = mod ([h-2/3  h  h-1/3], 1);

  ## factor s*v -> f
  f = repmat (s .* v, 1, 3);

  ## add s*v*hue-function to rgb map
  rgb += f .* (6 * (hue < 1/6) .* hue
               + (hue >= 1/6 & hue < 1/2)
               + (hue >= 1/2 & hue < 2/3) .* (4 - 6 * hue));

  rgb = colorspace_conversion_revert (rgb, cls, sz, is_im, is_nd, is_int);

endfunction

## Test pure colors
%!assert (hsv2rgb ([0 0 1]), [1 1 1])
%!assert (hsv2rgb ([1 1 0]), [0 0 0])
%!assert (hsv2rgb ([0 1 1]), [1 0 0])
%!assert (hsv2rgb ([1 1 1]), [1 0 0])
%!assert (hsv2rgb ([1/3 1 1]), [0 1 0])
%!assert (hsv2rgb ([2/3 1 1]), [0 0 1])

%!assert (hsv2rgb ([0 0.5 0.5]), hsv2rgb ([1 0.5 0.5]))

## Not Matlab compatible (Matlab would return [1/2  1/12  1/12]) but
## it's also invalid input.  This is, however, the same output as python
## colorsys module.
%!assert (hsv2rgb ([0.5 -0.5 0.5]), [0.75 0.5 0.5])

## Not Matlab compatible.  Matlab returns NaN when hue is outside the
## [0 1] range.  But since it's an angle, we can manage it.
%!assert (hsv2rgb ([0 0.5 0.5]), hsv2rgb ([2 0.5 0.5]))
%!assert (hsv2rgb ([0.5 0.5 0.5]), hsv2rgb ([2.5 0.5 0.5]), eps)

%!test
%! hsv_map = rand (64, 3);
%! assert (rgb2hsv (hsv2rgb (hsv_map)), hsv_map, 1e-6);

%!test
%! hsv_img = rand (64, 64, 3);
%! assert (rgb2hsv (hsv2rgb (hsv_img)), hsv_img, 1e-6);

## support sparse input
%!assert (hsv2rgb (sparse ([0 0 1])), sparse ([1 1 1]))
%!assert (hsv2rgb (sparse ([0 1 1])), sparse ([1 0 0]))
%!assert (hsv2rgb (sparse ([1 1 1])), sparse ([1 0 0]))

## Test input validation
%!error hsv2rgb ()
%!error hsv2rgb (1,2)
%!error <invalid data type> hsv2rgb ({1})
%!error <HSV must be a colormap or HSV image> hsv2rgb (ones (2,2))

## Test ND input
%!test
%! hsv = rand (16, 16, 3, 5);
%! rgb = zeros (size (hsv));
%! for i = 1:5
%!   rgb(:,:,:,i) = hsv2rgb (hsv(:,:,:,i));
%! endfor
%! assert (hsv2rgb (hsv), rgb)